1. Field of the Invention
The present invention relates to a workload calculation apparatus and method for an internal combustion engine and an engine control unit for calculating a workload parameter indicative of the workload of the internal combustion engine.
2. Description of the Prior Art
Conventionally, as this type of workload calculation apparatus, one described, for example, in Laid-open Japanese Patent Application No. 2006-52647 is known. This workload calculation apparatus calculates an indicated mean effective pressure of the internal combustion engine as follows. A piezoelectric-element-type in-cylinder pressure sensor, which is attached to an ignition plug, detects an in-cylinder pressure in a combustion chamber of the internal combustion engine. Then, the output of the in-cylinder pressure sensor and the volume data on the rate of change in volume of the combustion chamber are developed in Fourier series to calculate the Fourier coefficients. Based on the Fourier coefficients thus calculated, an indicated mean effective pressure of the internal combustion engine is calculated.
In general, an in-cylinder pressure sensor outputs, not the in-cylinder pressure itself, but a detection signal indicative of the amount of change in the in-cylinder pressure. Therefore, when this type of in-cylinder pressure sensor is used to calculate an indicated mean effective pressure as described above, the output of the in-cylinder pressure sensor must be integrated by, for example, a charge amplifier to obtain data on the in-cylinder pressure, and then the indicated mean effective pressure must be calculated based on the Fourier coefficient of the in-cylinder pressure data thus obtained. In this case, however, since the aforementioned type of in-cylinder pressure sensor has the characteristics of changing its output in accordance with temperature, the changes of the output in accordance with temperature accumulate as a result of the integral, leading to a great drift in the in-cylinder pressure data. As a result, the indicated mean effective pressure cannot possibly be calculated accurately.
The charge amplifier used may have the function of resetting the integral value at a predetermined timing during one combustion cycle of the internal combustion engine. However, in this case, because the integral is performed from combustion cycle to combustion cycle, the drift in the in-cylinder pressure data is not eliminated, thus also making an accurate calculation of the indicated mean effective pressure impossible. Also, a contiguous sequence of in-cylinder pressure data is inevitably interrupted before and after the process of resetting the integral value. As a result, the proper frequency characteristics of the in-cylinder pressure data cannot be obtained. Because of this, an appropriate Fourier coefficient cannot be evaluated, which in turn causes a reduction in accuracy of the indicated mean effective pressure.
Further, as an in-cylinder pressure detection apparatus for compensating for the aforementioned drift in the in-cylinder pressure data (hereinafter referred to as “in-cylinder pressure drift”), one known one is the one disclosed, for example, in Laid-open Japanese Patent Application No. 2005-256775 and Laid-open Japanese Patent Application No. 2005-315208. In these in-cylinder pressure detection apparatuses, the in-cylinder pressure drift in a current combustion cycle is calculated and then the in-cylinder pressure data in the next combustion cycle is corrected on the basis of the in-cylinder pressure drift thus calculated to compensate for the in-cylinder pressure drift. However, since these in-cylinder pressure detection apparatuses correct the in-cylinder pressure data in the next combustion cycle on the basis of the in-cylinder pressure drift in a current combustion cycle, these detection apparatuses are not capable of adequately compensating for the in-cylinder pressure drift in a transition period in which the in-cylinder pressure peak value greatly varies from combustion cycle to combustion cycle. In consequence, due to the use of the in-cylinder pressure data thus obtained, the indicated mean effective pressure cannot be calculated accurately.